Release 1.0 is shipped. Testing is very extensive and a huge list of bugs are found. The most critical ones are fixed, the rest are scheduled for Patch 1.0. The experienced part of the team moves onto their next project or takes a vacation. Now a load of new people are handed copies of Release 1.0 and assigned a bug. Most of them will manage, but a minority of them will make chages with severe side effects - e.g. their code will corrupt the stack or heap. They module test, missing the corruption and check the code in.
So now Patch 1.0 contains a lot of fixes, some very badly coded. Possibly they will cause problems on their own, or possibly when combined. There are bugs that were missed in the big release too. A lot of the new people will get assigned off the project. Usually the amount of system testing on pathces is not as much as Release 1.0
The other issue is that the commercial pressure on the company is dropping - bugs introduced by a patch when people have already paid are less serious commercially than bugs at release when they're still thinking about paying.
So it's quite possible that updates will actually make a product worse.
I use a nasty application at work in Java. Some versions only work with JRE 1.4 and some only work with JRE 1.5. They even have a message if the wrong version is installed. Actually.Net is just as shitty. Lookout, an excellent Outlook email indexer only works with.Net 1.1. If you have 2.0 installed it will crash
There are various hackarounds for this, like using a manifest to force Outlook to load.Net 1.1 or even hacking the binary of Lookout, but on my work machine I just uninstalled.Net 2.0 because I don't use anything that needs it.
Of course the best option would have been for the Lookout people to release an updated binary of Lookout to fix the bug that makes it require an old version of.Net. But there's no chance of that because Microsoft bought the company and took down the website. Lookout is now Microsoft Desktop Search, which I probably should be using instead.
I wonder what would happen if you made a "Pirate Doubloons" eCash site. You wouldn't mention that it was useful for legally dubious stuff, but you would point list among its features complete untraceability.
Basically like The Pirate Bay but for untraceable ecommerce.
I think Wolfram is a smart guy. The problem is that he probably has enough money now that he doesn't really have to listen to criticism. He works for his own company in which he is a majority shareholder so he can't get fired. His ideas (e.g. A new kind of science) are published by normal publishers, not peer reviewed journals. In every professional interaction in his life he's made sure he's in control, i.e. no one can tell him that he's talking out of his ass.
The problem is that humans - especially smart ones - have an enormous capacity for self delusion. Back when he was a physicist the people he worked with, the peer reviewers of the journals he submitted papers to and the people he wrote grant application to would act to keep him relatively grounded. Now even if he wrote wild speculation or even complete nonsense he's rich and famous enough to find someone to publish it, and they'll probably sell enough copies to do it again. Even if they didn't he's probably set for life with the money he got from Mathematica.
So he's free, but that sort of freedom is a very dangerous thing if you actually want to achieve anything intellectually.
Gmail does "selective degradation" really well. E.g. if you load gmail over a slow VPN over wireless connection it says "This site is taking longer to load than normal, would you like to try the Basic Html version or wait longer". Also you can choose basic html (i.e. less ajax and css) as your default view.
Basic HTML is quite usable these days - it even does email address autocompletion on Opera. So it can use ajax but it presumably doesn't depend of it. In a way it's a bit like a well written application which can use new features if they are present but run without them on downlevel systems.
I've worked with a lot of people that were dumb enough that I'd have no problem eating them. Not that I forsee a situation where that would be necessary, as I've tried to explain to HR.
The problem with that is that you end up with a small number of big companies owning patents which end up covering everything. If a patent war starts, they end up cross licensing.
Then if a new company tries to join the market they end up getting obliterated. So defensive patents tend to act as a barrier to new entrants, at least in the US. Of course outside the US it doesn't tend to work like that. Still if you want to see your stuff in the US you need to negotiate a fee.
Don't most people use the new Uninifieud Engluish spellings of colouur and alumininiuum, rather than the American specifics like color/aluminum or the British specificis like colour/aluminium?
Sorry, nobody informed you, but the Plain Text forces were soundly defeated in the great email wars of 1996.
You are that crazy japanese solider still fighting WWII in that episode of Gillian's Island.
True, and their leader General Neckbeard was subsequently forced to resign following demonstrations where students tore down statues of RMS, burned cheeto stained wolfshirts, deleted copies of Pine and installed Eudora.
Slashdot Mirror
Usually it's like this
Release 1.0 is shipped. Testing is very extensive and a huge list of bugs are found. The most critical ones are fixed, the rest are scheduled for Patch 1.0. The experienced part of the team moves onto their next project or takes a vacation. Now a load of new people are handed copies of Release 1.0 and assigned a bug. Most of them will manage, but a minority of them will make chages with severe side effects - e.g. their code will corrupt the stack or heap. They module test, missing the corruption and check the code in.
So now Patch 1.0 contains a lot of fixes, some very badly coded. Possibly they will cause problems on their own, or possibly when combined. There are bugs that were missed in the big release too. A lot of the new people will get assigned off the project. Usually the amount of system testing on pathces is not as much as Release 1.0
The other issue is that the commercial pressure on the company is dropping - bugs introduced by a patch when people have already paid are less serious commercially than bugs at release when they're still thinking about paying.
So it's quite possible that updates will actually make a product worse.
I use a nasty application at work in Java. Some versions only work with JRE 1.4 and some only work with JRE 1.5. They even have a message if the wrong version is installed. Actually .Net is just as shitty. Lookout, an excellent Outlook email indexer only works with .Net 1.1. If you have 2.0 installed it will crash
There are various hackarounds for this, like using a manifest to force Outlook to load .Net 1.1 or even hacking the binary of Lookout, but on my work machine I just uninstalled .Net 2.0 because I don't use anything that needs it.
Of course the best option would have been for the Lookout people to release an updated binary of Lookout to fix the bug that makes it require an old version of .Net. But there's no chance of that because Microsoft bought the company and took down the website. Lookout is now Microsoft Desktop Search, which I probably should be using instead.
Because the needs of the many outweigh the needs of the few.
Duh.
The summary is about graceful degradation. TFA is OFFTOPIC here.
I wonder what would happen if you made a "Pirate Doubloons" eCash site. You wouldn't mention that it was useful for legally dubious stuff, but you would point list among its features complete untraceability.
Basically like The Pirate Bay but for untraceable ecommerce.
I think Wolfram is a smart guy. The problem is that he probably has enough money now that he doesn't really have to listen to criticism. He works for his own company in which he is a majority shareholder so he can't get fired. His ideas (e.g. A new kind of science) are published by normal publishers, not peer reviewed journals. In every professional interaction in his life he's made sure he's in control, i.e. no one can tell him that he's talking out of his ass.
The problem is that humans - especially smart ones - have an enormous capacity for self delusion. Back when he was a physicist the people he worked with, the peer reviewers of the journals he submitted papers to and the people he wrote grant application to would act to keep him relatively grounded. Now even if he wrote wild speculation or even complete nonsense he's rich and famous enough to find someone to publish it, and they'll probably sell enough copies to do it again. Even if they didn't he's probably set for life with the money he got from Mathematica.
So he's free, but that sort of freedom is a very dangerous thing if you actually want to achieve anything intellectually.
http://www.wolframalpha.com/input/?i=(sinc+(x)+*+sinc+(y))+
I'm sure you could do better if you had more time.
if a small batch of users have to take a performance hit to improve the experience in the end for all users, isn't that a positive thing?
Didn't Jesus say "the needs of the many outweigh the needs of the few"?
Gmail does "selective degradation" really well. E.g. if you load gmail over a slow VPN over wireless connection it says "This site is taking longer to load than normal, would you like to try the Basic Html version or wait longer". Also you can choose basic html (i.e. less ajax and css) as your default view.
Basic HTML is quite usable these days - it even does email address autocompletion on Opera. So it can use ajax but it presumably doesn't depend of it. In a way it's a bit like a well written application which can use new features if they are present but run without them on downlevel systems.
That model is a beauty. Real quality. Mind you might want to take out the extended service plan on that model sir, repairs can be expensive.
Hmm, good point. But you accept that the genetically unclean must be prevented from occupying sensitive posts for the greater good of the race, right?
His book lernin' is pretty selective, he's managed to ignore the downside to Che completely.
http://www.huffingtonpost.com/johann-hari/this-endless-myth-making_b_151217.html
Whooee boys! We done caught ourselves a commie!
Your problem is that you leave them alive.
Stalin used to say to his minions that "I cantell when a man is going to betray me"
I'd hit it.
Your wife, Angelina and the rotten turnip. Quite possibly all at once.
YEAAAHHH!
I've worked with a lot of people that were dumb enough that I'd have no problem eating them. Not that I forsee a situation where that would be necessary, as I've tried to explain to HR.
The problem with that is that you end up with a small number of big companies owning patents which end up covering everything. If a patent war starts, they end up cross licensing.
Then if a new company tries to join the market they end up getting obliterated. So defensive patents tend to act as a barrier to new entrants, at least in the US. Of course outside the US it doesn't tend to work like that. Still if you want to see your stuff in the US you need to negotiate a fee.
I've always liked Courier White. E.g. look at this text
Far more readable, than regular Courier don't you think?
Don't most people use the new Uninifieud Engluish spellings of colouur and alumininiuum, rather than the American specifics like color/aluminum or the British specificis like colour/aluminium?
Sorry, nobody informed you, but the Plain Text forces were soundly defeated in the great email wars of 1996.
You are that crazy japanese solider still fighting WWII in that episode of Gillian's Island.
True, and their leader General Neckbeard was subsequently forced to resign following demonstrations where students tore down statues of RMS, burned cheeto stained wolfshirts, deleted copies of Pine and installed Eudora.
That abuse post is kind of clever
The subject was
1.2.3 ‮(lufthgisnI ,5:erocS)
U+8238 is combining cyrillic millions, it flips the text direction from that point on the line on.
Now when posted it was modded Offtopic, but what you see is
1.2.3 (cipotffO, 1-:erocS) (Score:5, Insightful)
Url Korrekt.
Sleep deprivation is a myth. Margaret Thatcher got by with four hours sleep a night, before she went mad.
Time travel probably.